Where to get parts for Hantek DSO scope?

[gahamb]

Hi all,

I have a Hantek DSO5102B which doesn't boot up. It just shows a blank screen and a couple of  buttons light. I checked the power supply voltages and they seem OK. Scoping the pins of various ICs such as the RAM I can't see pulses anywhere, however I can see that the 12 MHz crystal is working. So it looks like the main SOC is not working. I have a USB to Serial interface on order which I'll use to see if any boot messages are showing up to double check if the system is actually booting.

So, assuming it is the main board that is broken, I started looking for a replacement (pn T0503717) first on the on Hantek site for service agents then Google, eBay etc., but couldn't find a supplier of the board (or even used scopes).

So does anyone know if and where I could get a replacement main board (and probable cost)? The one I have is HW version 1.00.5 (which I understand was buggy and only produced for a short time). 

I also understand that this scope was produced under other manufacturers such as Tekway's DST1102B and Voltcraft's DSO3062D (60MHz version). So mother boards for these scopes my be usable in the Hantek. However, I have the same problem in that I can't find any suppliers for parts for these scopes either.

Thanks,
Graham

[gahamb]

I asked a number of suppliers of this scope whether they had spares and what would it cost for a main board (including Hantek themselves).

James from Test Equipment Pty Ltd just replied that it would cost $750 plus GST   . Usual story about cost of spare parts - sometimes cheaper to get a whole new item.

Regards,
Graham

[Fungus]

So does anyone know if and where I could get a replacement main board (and probable cost)?

From another Hantek scope?

Seriously: Stocking and supplying replacement parts for all their models for any reasonable length of time will make the PCB cost as much as a new 'scope (maybe more!) so it simply won't ever happen.

[NCG]

Just to be sure, where you did check the voltages? On PSU side or on motherboard? 2 white power connectors that go to main board - those oxidize badly and you might be missing some of supply voltages still.

[gahamb]

Hi NCG,

Yes, I checked the voltages on the PSU and on the motherboard. I also went through the various test points on the PCB guided by the schematics done by tinhead (great work!) and checked the output of the various regulators generating the FPGA voltages. They were all in spec.

Thanks,
Graham

[gahamb]

Seriously: Stocking and supplying replacement parts for all their models for any reasonable length of time will make the PCB cost as much as a new 'scope (maybe more!) so it simply won't ever happen.

The local distributor doesn't keep any stock here (or very minimal levels), they order in as required so their cost is just what the factory charges them plus their (distrib) markup.

I was hoping that since these scopes had been around for many years now that there may be some secondhand market in them, but there is none around and the historical sales of used ones on eBay is almost non existent . So either they are all still working and no-one wants to upgrade, or when they do fail they are simply chucked out and not put up for sale for parts. I suspect the latter.

Graham

[supercilious]

Just a heads-up: I've found on my own Hantek 5102P that the cause of a similar failure was the flash chip wearing out and the SoC no longer reading back a valid filesystem (UBIFS). The scope simply died during boot.

You might find that replacing the chip and flashing it with a fresh copy of the firmware image from a working scope (remember to zero out the UBIFS OoB data for wear leveling).

[gahamb]

Just a heads-up: I've found on my own Hantek 5102P that the cause of a similar failure was the flash chip wearing out and the SoC no longer reading back a valid filesystem (UBIFS). The scope simply died during boot.

Thanks for the heads up about your experience, I'll investigate your suggestion. Do you know offhand whether the 5102P has a similar architecture and PCB/component list to the 5102B? 

One thing about my situation is that I cannot find a clock signal anywhere on the board except for the pins of the xtal. I would assume that even if the flash chip was faulty, the SOC should still be propagating the clocks. So It looks like the SOC is faulty (or has a bad connection).

You might find that replacing the chip and flashing it with a fresh copy of the firmware image from a working scope (remember to zero out the UBIFS OoB data for wear leveling).

Do you have a link to a document describing the procedure to re-load the firmware?

Regards,
Graham

[supercilious]

I've never owned a 5102B, and I'm not sure where my 5102P is exactly at the moment, but from what I recall, the only difference is the sample memory - the 5102P has only its internal 40KS whereas the 5102B has a 1MS memory they are otherwise identical. I think they even have the same board, but both memory pads are unpopulated on the P, 1 is unpopulated on the B and there is another version with both populated with 2MS memory. Not 100% sure though - it might be just 1 pad and the memory difference is the specific memory chip rather than 1 or 2 of them.

In my case, the serial pins on the board gave some output from the bootloader showing that it attempted to boot but the kernel failed to mount the filesystem containing the scope software and crashed due to the kernel being unable to exec() the /init binary. I was able to get a firmware image from Hantek via email that could be flashed via the bootloader to replace the corrupted image. The flash is partitioned into various small block devices /dev/mtd[1-6] and 4,5,6 were corrupt on mine and required reflashing. I expect that the others being corrupt would mean that you'd have to flash externally rather than via the serial because the bootloader itself wouldn't run.

[gahamb]

Well, today I turned the scope on and it booted up no problems . I played with the menus a bit, turned it off and on again a few times, checked a probe calibration waveform and the controls - no problems. After a couple of minutes, I decided to see if I could update the firmware, but before I could start, the menus locked up and now it won't boot at all .

So whatever is wrong, it doesn't seem like a catastrophic failure. Possibly something not being reset correctly on power up, or just an intermittent problem.

I'm still waiting for the USB to serial interface board to arrive to see if there are any meaningful messages being sent out before it dies.

Graham

[PA4TIM]

Did you check ripple on the psu rails ? I repaired a Fluke 123 that had so much ripple that the DC voltage looked not bad. And that caused all sorts of strange faults.
I had a DSO here a while back that also had bootproblems. Turned out the "psu watchdog" was confused because psu problems. The psu looked fine but that was without the load of a working scope. As soon as it started booting the watchdog shut it down.

Next thing that I would check is all solderjoints (under a microscope).  I see a lot of problems with bad solderjoints.

I'm not surprised you can not get parts for these DSO's.  They are build with just one goal, to be as cheap as possible, last the warranty time and after that..they don't care  ;-) at least they give you a reply. Fluke can learn from that 

[BravoV]

Did you check ripple on the psu rails ?

The OP needs "ANOTHER" healthy & working scope to do that, which is unclear whether the op owns it, I doubt it, cmiiw. 

[PA4TIM]

One thing about my situation is that I cannot find a clock signal anywhere on the board except for the pins of the xtal. I would assume that even if the flash chip was faulty, the SOC should still be propagating the clocks. So It looks like the SOC is faulty (or has a bad connection).

Because of this I assumed he had an other scope.

[gahamb]

Because of this I assumed he had an other scope.

Yes, I have a Tek 2232 (which has it's own problems, but that's another story). So I can and will check the PSU ripple.

Regards,
Graham

[gahamb]

I had a DSO here a while back that also had bootproblems. Turned out the "psu watchdog" was confused because psu problems. The psu looked fine but that was without the load of a working scope. As soon as it started booting the watchdog shut it down.

Sounds a bit like my problem. Most of the time when I power it on there is no response and there are no clocks. Do you know if there is such a "PSU watchdog" on this board or is it a separate chip (I don't see one marked anywhere) or would it be part of the SOC?
 

Next thing that I would check is all solderjoints (under a microscope).  I see a lot of problems with bad solderjoints.

I had a look with a magnifying glass (I don't have a microscope) but couldn't see anything obvious. I also tried flexing and tapping the board to try to induce the fault mechanically but couldn't. The problem is that it starts up so infrequently, it takes a while to try various things.

Regards,
Graham

[Fungus]

I'd start by looking at the power supply.

If it boots for a while then the PCB is probably OK.


Put a load on the power supply and watch the voltage levels over time as it warms up.

[ebclr]

Those China low-cost products does not have any support ( 90% of then )

Just use while is ok, any defect they will replace when on warranty or buy a new one if is no more on warranty

Forget things like the schematic . service manual and spare parts, they will simply ignore you or will price higher than the full instrument.

It's the cheap and detachable philosophy

Some exceptions to this rule but very very few brands

[supercilious]

While I am in no way advocating for a Hantek scope (quite the contrary, I would suggest people avoid them and get something else), I must say that it was my first ever scope and when I suffered the NAND error, Hantek staff were very helpful in diagnosing and fixing it.

I didn't really know about these forums so I didn't even know about the serial interface on the board or anything. They (Hantek) provided detailed instructions to help me figure out what was wrong and then provided a firmware image and flashing instructions and tools to recover it. I know that Chinese equipment manufacturers have a poor reputation here, but all my experiences have been positive.

I think that the positive experiences rarely get mentioned here because only the people who are not satisfied by the support they receive bother to come to forums like this and complain, the rest of us just get on with our lives and our now working instruments making places like this an echo chamber for bad experiences.

[NCG]

Well, if it sometimes starts I would still bet it is just the 2 power connectors oxidizing - you get voltage drop spikes with some instructions and CPU goes to HLT. I just soldered gold plated spikes (from 2.54mm array) there instead of originals and slightly scraped the other side. That was it, no more crashes and black screen starts.